Numerical estimation of the permeability of granular soils using the DEM and LBM or FFT-based fluid computation method

Numerical packings of spheres with uniform grain size distribution and maximum to minimum diameter ratio up to 15 are generated using the Discrete Element Method (DEM). Two numerical methods are used to compute their permeability by homogenization: the Lattice Boltzmann Method (LBM) and a Fast Fouri...

Full description

Saved in:
Bibliographic Details
Published in:Granular matter Vol. 25; no. 3; p. 53
Main Authors: Nguyen, Ngoc Son, Bignonnet, François
Format: Journal Article
Language:English
Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.08.2023
Springer Nature B.V
Springer Verlag
Subjects:
Civil Engineering
Complex Fluids and Microfluidics
Computing time
Discrete element method
Discretization
Engineering Fluid Dynamics
Engineering Sciences
Engineering Thermodynamics
Fast Fourier transformations
Fluids mechanics
Foundations
Fourier transforms
Geoengineering
Grain size distribution
Géotechnique
Heat and Mass Transfer
Hydraulics
Industrial Chemistry/Chemical Engineering
Materials Science
Mechanics
Numerical methods
Original Report
Permeability
Physics
Physics and Astronomy
Soft and Granular Matter
Soil permeability
Fast Fourier transform (FFT)
Homogenization
Discrete element method (DEM)
Lattice boltzmann method (LBM)
Permeability
Granular materials
Lattice Boltzmann method (LBM)
Discrete Element Method (DEM)
Fast Fourier Transform (FFT)
ISSN:1434-5021, 1434-7636
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Numerical packings of spheres with uniform grain size distribution and maximum to minimum diameter ratio up to 15 are generated using the Discrete Element Method (DEM). Two numerical methods are used to compute their permeability by homogenization: the Lattice Boltzmann Method (LBM) and a Fast Fourier Transform (FFT) based method. The results given by both methods are shown to be consistent with semi-analytical and experimental results. For an identical discretization grid, the FFT method has the lowest memory and computational time requirements. The LBM is more accurate for coarse to moderately fine discretizations, while the FFT method converges linearly with the voxel size h with a relative discretization error below 1.5 times h / D 25 , where D 25 is the 25% passing by mass grain diameter. The issue of the variability of the permeability computed on finite sized samples is determined either directly by many realizations of similar random samples or indirectly by a faster filtering method on a single sample. Both methods yield similar results and indicate that a Representative Volume Element (RVE) size greater than 7 D 40 guarantees a variability of permeability below 5%. Graphic abstract
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1434-5021
1434-7636
DOI:10.1007/s10035-023-01330-1